Simplification of significance map coding

1. A method for encoding video data comprising: receiving, by a computing device, an array of transform coefficients corresponding to a luma component or a chroma component of the video data; selecting, by the computing device, using a shared L1 context derivation process that applies to both the luma component and the chroma component, L1 context models wherein the L1 context models are selected from a shared set of L1 context models that apply to both the luma component and the chroma component and the selecting includes, for each sub-block in the array: determining a significance value of a right neighbor sub-block of the sub-block and a significance value of a bottom neighbor sub-block of the sub-block, and selecting one of a first L1 context model or a second L1 context model for encoding the significance value of the sub-block based only on the significance value of the right neighbor sub-block and the significance value of the bottom neighbor sub-block of the sub-block, wherein the first L1 context model is selected if neither the significance value of the right neighbor sub-block nor the significance value of the bottom neighbor sub-block is equal to one and the second L1 context model is selected if the significance value of the right neighbor sub-block or the significance value of the bottom neighbor sub-block is equal to one; encoding, by the computing device, an L1 significance map that includes L1 significance values for sub-blocks in the array using the context models; selecting, by the computing device, using a shared L0 context derivation process that applies to both the luma component and the chroma component, L0 context models wherein the L0 context models are selected from a shared set of L0 context models that apply to both the luma component and the chroma component and the selecting includes, for each sub-block in the array: partitioning the array into a first region and a second region using a shared partitioning process, wherein the first region comprises a DC coefficient of the array and the second region comprises transform coefficients in the array other than the DC coefficient, selecting, for each sub-block having a non-zero significance value, when the sub-block is in the first region, a first L0 context model for encoding the significance value of the transform coefficients of the sub-block, selecting, for each sub-block having a non-zero significance value, when the sub-block is in the second region, one of a second L0 context model, a third L0 context model, a fourth L0 context model or a fifth L0 context model for encoding the significance value of a transform coefficient of the sub-block having a non-zero significance value based on significance values of four neighbor transform coefficients, wherein the second L0 context model is selected if there are no non-zero significance values among the four neighbor transform coefficients, the third L0 context model is selected if there is one non-zero significance value among the four neighbor transform coefficients, the fourth L0 context model is selected if there are two non-zero significance values among the four neighbor transform coefficients, and the fifth L0 context model is selected if there are more than two non-zero significance value among the four neighbor transform coefficients; and encoding, by the computing device, an L0 significance map that includes L0 significance values for sub-blocks in the array using the L0 context models.

2. A method for decoding video data comprising: receiving, by a computing device, a bitstream of compressed data, the compressed data including an array of transform coefficients that was previously encoded, the array corresponding to a luma component or a chroma component of the video data; selecting, by the computing device, using a shared L1 context derivation process that applies to both the luma component and the chroma component, L1 context models wherein the L1 context models are selected from a shared set of L1 context models that apply to both the luma component and the chroma component and the selecting includes, for each sub-block in the array: determining a significance value of a right neighbor sub-block of the sub-block and a significance value of a bottom neighbor sub-block of the sub-block, and selecting one of a first L1 context model or a second L1 context model for decoding the significance value of the sub-block based only on the significance value of the right neighbor sub-block and the significance value of the bottom neighbor sub-block of the sub-block, wherein the first L1 context model is selected if neither the significance value of the right neighbor sub-block nor the significance value of the bottom neighbor sub-block is equal to one and the second L1 context model is selected if the significance value of the right neighbor sub-block or the significance value of the bottom neighbor sub-block is equal to one; decoding, by the computing device, an L1 significance map that includes L1 significance values for sub-blocks in the array using the context models; selecting, by the computing device, using a shared L0 context derivation process that applies to both the luma component and the chroma component, L0 context models wherein the L0 context models are selected from a shared set of L0 context models that apply to both the luma component and the chroma component and the selecting includes, for each sub-block in the array: partitioning the array into a first region and a second region using a shared partitioning process, wherein the first region comprises a DC coefficient of the array and the second region comprises transform coefficients in the array other than the DC coefficient, selecting, for each sub-block having a non-zero significance value, when the sub-block is in the first region, a first L0 context model for decoding the significance value of the transform coefficients of the sub-block, selecting, for each sub-block having a non-zero significance value, when the sub-block is in the second region, one of a second L0 context model, a third L0 context model, a fourth L0 context model or a fifth L0 context model for decoding the significance value of a transform coefficient of the sub-block having a non-zero significance value based on significance values of four neighbor transform coefficients, wherein the second L0 context model is selected if there are no non-zero significance values among the four neighbor transform coefficients, the third L0 context model is selected if there is one non-zero significance value among the four neighbor transform coefficients, the fourth L0 context model is selected if there are two non-zero significance values among the four neighbor transform coefficients, and the fifth L0 context model is selected if there are more than two non-zero significance value among the four neighbor transform coefficients; and decoding, by the computing device, an L0 significance map that includes L0 significance values for sub-blocks in the array using the L0 context models.

3. A non-transitory computer-readable medium including program code executable by a processor for encoding video data, wherein the program code, when executed by the processor, causes the processor to perform operations, the operations comprising: receiving an array of transform coefficients corresponding to a luma component or a chroma component of the video data; selecting, using a shared L1 context derivation process that applies to both the luma component and the chroma component, L1 context models wherein the L1 context models are selected from a shared set of L1 context models that apply to both the luma component and the chroma component and the selecting includes, for each sub-block in the array: determining a significance value of a right neighbor sub-block of the sub-block and a significance value of a bottom neighbor sub-block of the sub-block, and selecting one of a first L1 context model or a second L1 context model for encoding the significance value of the sub-block based only on the significance value of the right neighbor sub-block and the significance value of the bottom neighbor sub-block of the sub-block, wherein the first L1 context model is selected if neither the significance value of the right neighbor sub-block nor the significance value of the bottom neighbor sub-block is equal to one and the second L1 context model is selected if the significance value of the right neighbor sub-block or the significance value of the bottom neighbor sub-block is equal to one; encoding an L1 significance map that includes L1 significance values for sub-blocks in the array using the context models; selecting using a shared L0 context derivation process that applies to both the luma component and the chroma component, L0 context models wherein the L0 context models are selected from a shared set of L0 context models that apply to both the luma component and the chroma component and the selecting includes, for each sub-block in the array: partitioning the array into a first region and a second region using a shared partitioning process, wherein the first region comprises a DC coefficient of the array and the second region comprises transform coefficients in the array other than the DC coefficient, selecting, for each sub-block having a non-zero significance value, when the sub-block is in the first region, a first L0 context model for encoding the significance value of the transform coefficients of the sub-block, selecting, for each sub-block having a non-zero significance value, when the sub-block is in the second region, one of a second L0 context model, a third L0 context model, a fourth L0 context model or a fifth L0 context model for encoding the significance value of a transform coefficient of the sub-block having a non-zero significance value based on significance values of four neighbor transform coefficients, wherein the second L0 context model is selected if there are no non-zero significance values among the four neighbor transform coefficients, the third L0 context model is selected if there is one non-zero significance value among the four neighbor transform coefficients, the fourth L0 context model is selected if there are two non-zero significance values among the four neighbor transform coefficients, and the fifth L0 context model is selected if there are more than two non-zero significance value among the four neighbor transform coefficients; and encoding an L0 significance map that includes L0 significance values for sub-blocks in the array using the L0 context models.

4. A non-transitory computer-readable medium have stored thereon program code executable by a processor for decoding video data, the program code comprising: receiving a bitstream of compressed data, the compressed data including an array of transform coefficients that was previously encoded, the array corresponding to a luma component or a chroma component of the video data; selecting, using a shared L1 context derivation process that applies to both the luma component and the chroma component, L1 context models wherein the L1 context models are selected from a shared set of L1 context models that apply to both the luma component and the chroma component and the selecting includes, for each sub-block in the array: determining a significance value of a right neighbor sub-block of the sub-block and a significance value of a bottom neighbor sub-block of the sub-block, and selecting one of a first L1 context model or a second L1 context model for decoding the significance value of the sub-block based only on the significance value of the right neighbor sub-block and the significance value of the bottom neighbor sub-block of the sub-block, wherein the first L1 context model is selected if neither the significance value of the right neighbor sub-block nor the significance value of the bottom neighbor sub-block is equal to one and the second L1 context model is selected if the significance value of the right neighbor sub-block or the significance value of the bottom neighbor sub-block is equal to one; decoding an L1 significance map that includes L1 significance values for sub-blocks in the array using the context models; selecting using a shared L0 context derivation process that applies to both the luma component and the chroma component, L0 context models wherein the L0 context models are selected from a shared set of L0 context models that apply to both the luma component and the chroma component and the selecting includes, for each sub-block in the array: partitioning the array into a first region and a second region using a shared partitioning process, wherein the first region comprises a DC coefficient of the array and the second region comprises transform coefficients in the array other than the DC coefficient, selecting, for each sub-block having a non-zero significance value, when the sub-block is in the first region, a first L0 context model for decoding the significance value of the transform coefficients of the sub-block, selecting, for each sub-block having a non-zero significance value, when the sub-block is in the second region, one of a second L0 context model, a third L0 context model, a fourth L0 context model or a fifth L0 context model for decoding the significance value of a transform coefficient of the sub-block having a non-zero significance value based on significance values of four neighbor transform coefficients, wherein the second L0 context model is selected if there are no non-zero significance values among the four neighbor transform coefficients, the third L0 context model is selected if there is one non-zero significance value among the four neighbor transform coefficients, the fourth L0 context model is selected if there are two non-zero significance values among the four neighbor transform coefficients, and the fifth L0 context model is selected if there are more than two non-zero significance value among the four neighbor transform coefficients; and decoding an L0 significance map that includes L0 significance values for sub-blocks in the array using the L0 context models.